Daron Acemoğlu
Last reviewed
Jun 8, 2026
Sources
6 citations
Review status
Source-backed
Revision
v1 · 1,606 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
Jun 8, 2026
Sources
6 citations
Review status
Source-backed
Revision
v1 · 1,606 words
Add missing citations, update stale details, or suggest a clearer explanation.
Daron Acemoglu (born September 3, 1967) is a Turkish-American economist and Institute Professor at the Massachusetts Institute of Technology (MIT). He is one of the most cited economists of his generation and a 2024 laureate of the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, shared with James A. Robinson and Simon Johnson. Best known for his work on how political and economic institutions shape long-run prosperity, Acemoglu has in recent years become a prominent and deliberately skeptical voice on the macroeconomics of artificial intelligence, arguing that the technology's near-term effects on productivity, growth, and wages are likely to be far more modest than industry forecasts suggest. [1][4]
Acemoglu was born in Istanbul, Turkey, to an Armenian family; his father, Kevork Acemoglu, was a commercial lawyer and his mother, Irma, was a school principal. He completed a BA in economics at the University of York in 1989, followed by an MSc in econometrics and mathematical economics at the London School of Economics (LSE) in 1990. He earned his PhD in economics from the LSE in 1992 with a thesis titled "Essays in Microfoundations of Macroeconomics: Contracts and Economic Performance," supervised by Kevin W. S. Roberts. [1]
After completing his doctorate, Acemoglu spent a year as a lecturer at the LSE before joining MIT in 1993 as an assistant professor. He received tenure in 1998 and was promoted to full professor in 2000. He held the Charles P. Kindleberger Professorship of Applied Economics from 2004 to 2010 and has been the Elizabeth and James Killian Professor of Economics since 2010. In July 2019, MIT named him an Institute Professor, the highest honor the university bestows on its faculty, reserved for a small number of scholars at any one time. [1]
Acemoglu is extraordinarily prolific, with a research output spanning growth theory, labor economics, political economy, network economics, and the economics of technology. By measures compiled by the RePEc bibliographic database, he has repeatedly ranked as the most cited economist of the past several decades. He is also a co-author, with Robinson, of widely read books that brought academic work on institutions to a general audience. [1]
Acemoglu's foundational research, much of it conducted with the political scientist James A. Robinson, argues that differences in national prosperity are driven primarily by institutions rather than by geography, culture, or short-run policy. The framework distinguishes "inclusive" institutions, which secure property rights, enforce contracts, and broaden political participation, from "extractive" institutions, which concentrate power and resources in a narrow elite. With Simon Johnson and Robinson, he used the differential mortality of European colonial settlers as a natural experiment to identify the causal effect of institutions on long-run income, a research design that became one of the most influential empirical strategies in development economics. [1]
This body of work was popularized in two best-selling books co-authored with Robinson: Why Nations Fail: The Origins of Power, Prosperity, and Poverty (2012) and The Narrow Corridor: States, Societies, and the Fate of Liberty (2019). Acemoglu also wrote the graduate textbook Introduction to Modern Economic Growth (2009) and co-authored Economic Origins of Dictatorship and Democracy (2006) with Robinson. [1]
A second major strand of Acemoglu's work, developed largely with Pascual Restrepo, reframes automation using a "task-based" model of production. In this framework, output is produced from a continuum of distinct tasks, each of which can be performed by either labor or capital. Automation expands the set of tasks performed by machines, generating a "displacement effect" that pushes workers out of those tasks and tends to lower the labor share of income. This can be offset by a "reinstatement effect" when new, labor-intensive tasks are created. Whether technology raises or lowers wages depends on the balance between these forces. [6]
Within this framework Acemoglu and Restrepo introduced the influential concept of "so-so automation" (or "so-so technologies"): automation that is just good enough to be adopted and to replace workers, yet not productive enough to generate large productivity gains. Because such technologies displace labor without delivering offsetting productivity growth, they can depress wages and the labor share while contributing little to aggregate output. Empirically, their 2020 study "Robots and Jobs" found that the spread of industrial robots in U.S. local labor markets reduced both employment and wages, and their 2022 paper "Tasks, Automation, and the Rise in U.S. Wage Inequality" attributed a substantial part of rising U.S. wage inequality since 1980 to automation that displaced workers specialized in routine tasks. [6]
Acemoglu has applied these long-standing ideas directly to the current wave of generative AI and large language models, making him one of the most influential skeptics in the economics of AI. His central argument is that the productive value of a technology is not predetermined: it depends on choices about whether AI is used to automate work and replace people or to augment workers by giving them new capabilities, information, and tasks. He worries that current incentives push developers toward automation and what he calls "so-so" applications that cut labor costs without large productivity benefits, with consequences for wages and inequality. [4]
These themes are developed at book length in Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity (2023), co-authored with Simon Johnson. Drawing on a millennium of technological history, the book argues that prosperity from new technology is not automatic but is shaped by who holds power and which uses of technology are pursued. Gains have historically been broadly shared only when institutions, worker bargaining power, and countervailing forces redirected innovation toward augmenting rather than displacing labor. Acemoglu and Johnson apply this lens to AI, cautioning that without deliberate choices and policy the technology could widen inequality and concentrate economic and political power. [5]
Acemoglu's most cited contribution to the AI debate is the paper "The Simple Macroeconomics of AI," issued as an NBER working paper in May 2024 and published in the journal Economic Policy in 2025. Using his task-based framework and an aggregation result known as Hulten's theorem, he estimates the macroeconomic upside of AI from two inputs: the share of work tasks that AI can affect and the average cost saving on those tasks. Drawing on a 2023 study estimating that roughly 20 percent of U.S. work tasks are exposed to AI, and on evidence that only a fraction of exposed tasks would be cost-effective to automate within a decade, he concludes that the gains will be real but small. [3]
His headline estimates are summarized below.
| Estimate (10-year horizon) | Acemoglu's figure |
|---|---|
| Increase in total factor productivity (TFP) | 0.53% to 0.66% total |
| Implied average annual TFP gain | roughly 0.05% per year |
| Increase in GDP | 1.1% to 1.6% total |
| Share of tasks exposed to AI | about 20% |
Acemoglu argues that even the upper bound of 0.66 percent is likely an overestimate, because the harder, more context-dependent tasks AI may eventually reach offer less clear-cut cost savings and lack objective performance measures. He further notes that some new AI-enabled activities, such as algorithmic manipulation or online deception, may carry negative social value, so that measured GDP gains could overstate true welfare improvements. On distribution, he finds little reason to expect AI to reduce labor income inequality and warns that it is likely to widen the gap between capital and labor income. These figures stand in sharp contrast to more bullish projections from some banks and consultancies, and the paper prompted public rebuttals from forecasters who had predicted much larger effects. [3][4]
Acemoglu has stressed that his skepticism is not dismissal. As he has put it, a productivity gain of around 0.5 percent over a decade is "better than zero," but he argues the central problem is that AI is being used "too much for automation and not enough for providing expertise and information to workers." He advocates redirecting AI development toward tools that complement human expertise, alongside policies on taxation, labor, and competition designed to spread the benefits more widely. [4]
Acemoglu received the John Bates Clark Medal in 2005, awarded to the American economist under 40 judged to have made the most significant contribution to economic thought. In 2024 he, James A. Robinson, and Simon Johnson were jointly awarded the Nobel Memorial Prize in Economic Sciences "for studies of how institutions are formed and affect prosperity," with the three sharing the award equally. He is a fellow of the National Academy of Sciences, the American Academy of Arts and Sciences, and the Econometric Society, among other honors. [1][2]
| Honor | Year |
|---|---|
| John Bates Clark Medal | 2005 |
| MIT Institute Professor | 2019 |
| Nobel Memorial Prize in Economic Sciences | 2024 |